End of training

README.md

@@ -66,7 +66,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/bcf688bed82a4bb7_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 10668bf8-55ca-4fc5-af41-66382e56c98d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 10668bf8-55ca-4fc5-af41-66382e56c98d
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 10668bf8-55ca-4fc5-af41-66382e56c98d
 
 This model is a fine-tuned version of [Qwen/Qwen2-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2-1.5B-Instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.3636
 
 ## Model description
 
@@ -128,14 +130,19 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
-- lr_scheduler_warmup_steps: 2
-- training_steps: 1
+- lr_scheduler_warmup_steps: 10
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0003 | 1    | 5.4379          |
+| 5.1103        | 0.0026 | 10   | 5.0619          |
+| 3.8694        | 0.0052 | 20   | 3.7605          |
+| 3.2688        | 0.0078 | 30   | 3.4719          |
+| 3.3279        | 0.0104 | 40   | 3.3841          |
+| 3.6121        | 0.0130 | 50   | 3.3636          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5754070d4a7e50751039f68195a18c3f78c80e45a64a7284fbcf5f080e4e3eff
+oid sha256:789f69c417108119281918c546ef735498f57259fffe586a3eb8aec859f7d944
 size 37070634